A Systematic Review on Automatic Insect Detection Using Deep Learning

Agriculture, Год журнала: 2023, Номер 13(3), С. 713 - 713

Опубликована: Март 19, 2023

Globally, insect pests are the primary reason for reduced crop yield and quality. Although pesticides commonly used to control eliminate these pests, they can have adverse effects on environment, human health, natural resources. As an alternative, integrated pest management has been devised enhance control, decrease excessive use of pesticides, output quality crops. With improvements in artificial intelligence technologies, several applications emerged agricultural context, including automatic detection, monitoring, identification insects. The purpose this article is outline leading techniques automated detection insects, highlighting most successful approaches methodologies while also drawing attention remaining challenges gaps area. aim furnish reader with overview major developments field. This study analysed 92 studies published between 2016 2022 insects traps using deep learning techniques. search was conducted six electronic databases, 36 articles met inclusion criteria. criteria were that applied classification, counting, written English. selection process involved analysing title, keywords, abstract each study, resulting exclusion 33 articles. included 12 classification task 24 task. Two main approaches—standard adaptable—for identified, various architectures detectors. accuracy found be influenced by dataset size, significantly affected number classes size. highlights two recommendations, namely, characteristics (such as unbalanced incomplete annotation) limitations algorithms small objects lack information about insects). To overcome challenges, further research recommended improve practices. should focus addressing identified ensure more effective management.

Язык: Английский

---

Human-Centered AI in smart farming: Towards Agriculture 5.0

IEEE Access, Год журнала: 2024, Номер 12, С. 62199 - 62214

Опубликована: Янв. 1, 2024

Язык: Английский

---

Plant disease recognition in a low data scenario using few-shot learning

Computers and Electronics in Agriculture, Год журнала: 2024, Номер 219, С. 108812 - 108812

Опубликована: Март 4, 2024

Plant disease is one of the major problems in agriculture. Diseases damage plants, reduce yields and lower quality produce. Traditional approaches to detecting plant diseases are usually based on visual inspection laboratory testing, which can be expensive time-consuming. They require trained pathologists as well specialised equipment. Several studies demonstrate that artificial intelligence (AI) methods produce promising results. However, AI generally data-hungry large annotated datasets, collection annotation such datasets a limiting factor. It often appears only small amount data available for certain types. Whereas performance typical drops significantly when they with inadequate data. This paper proposes novel few-shot learning (FSL) method detect alleviate scarcity problem. The proposed uses few five images per class machine process. Our state-of-the-art FSL pipeline called pre-training, meta-learning, fine-tuning (PMF), integrated feature attention (FA) module; we call overall PMF+FA. FA module emphasises discriminative parts image reduces impact complicated backgrounds undesired objects. We used ResNet50 Vision Transformers (ViT) learner. Two publicly were repurposed meet requirements. thoroughly evaluated PlantDoc dataset, contains samples field environments complex unwanted PMF+FA ViT achieved an average accuracy 90.12% recognition. results consistently outperforms baseline PMF. also highlight using generates better than diagnosing implementations computationally efficient, taking 1.11 0.57 ms evaluate test set respectively. high throughput high-quality training dataset indicate technique real-time detection digital farming systems.

Язык: Английский

---

Harnessing quantum computing for smart agriculture: Empowering sustainable crop management and yield optimization

Computers and Electronics in Agriculture, Год журнала: 2024, Номер 218, С. 108680 - 108680

Опубликована: Фев. 10, 2024

Язык: Английский

---

PestLite: A Novel YOLO-Based Deep Learning Technique for Crop Pest Detection

Agriculture, Год журнала: 2024, Номер 14(2), С. 228 - 228

Опубликована: Янв. 31, 2024

Timely and effective pest detection is essential for agricultural production, facing challenges such as complex backgrounds a vast number of parameters. Seeking solutions has become pressing matter. This paper, based on the YOLOv5 algorithm, developed PestLite model. The model surpasses previous spatial pooling methods with our uniquely designed Multi-Level Spatial Pyramid Pooling (MTSPPF). Using lightweight unit, it integrates convolution, normalization, activation operations. It excels in capturing multi-scale features, ensuring rich extraction key information at various scales. Notably, MTSPPF not only enhances accuracy but also reduces parameter size, making ideal models. Additionally, we introduced Involution Efficient Channel Attention (ECA) attention mechanisms to enhance contextual understanding. We replaced traditional upsampling Content-Aware ReAssembly FEatures (CARAFE), which enable achieve higher mean average precision detection. Testing dataset showed improved while reducing size. mAP50 increased from 87.9% 90.7%, count decreased 7.03 M 6.09 M. further validated using IP102 dataset, other hand, conducted comparisons mainstream Furthermore, visualized targets. results indicate that provides an solution real-time target pests.

Язык: Английский

---

Detection of citrus black spot disease and ripeness level in orange fruit using learning-to-augment incorporated deep networks

Ecological Informatics, Год журнала: 2022, Номер 71, С. 101829 - 101829

Опубликована: Сен. 24, 2022

Язык: Английский

---

Research on CBF-YOLO detection model for common soybean pests in complex environment

Computers and Electronics in Agriculture, Год журнала: 2023, Номер 216, С. 108515 - 108515

Опубликована: Дек. 19, 2023

Язык: Английский

---

Improved Pest-YOLO: Real-time pest detection based on efficient channel attention mechanism and transformer encoder

Ecological Informatics, Год журнала: 2023, Номер 78, С. 102340 - 102340

Опубликована: Окт. 20, 2023

Язык: Английский

---

Field detection of small pests through stochastic gradient descent with genetic algorithm

Computers and Electronics in Agriculture, Год журнала: 2023, Номер 206, С. 107694 - 107694

Опубликована: Фев. 14, 2023

Язык: Английский

---

Application of machine learning in automatic image identification of insects - a review

Ecological Informatics, Год журнала: 2024, Номер 80, С. 102539 - 102539

Опубликована: Фев. 23, 2024

Fast and reliable identification of insect species is crucial for pest management, animal quarantine, effective utilization resources. Traditional morphological classification time-consuming laborious, while automatic image techniques based on machine learning (ML) can greatly improve efficiency. ML a promising approach the identification, including traditional (TML) deep (DL). This review outlines process TML/DL. We highlighted methods acquisition, preprocessing, segmentation, detection. The applications various orders are summarized discussed, with focus Coleoptera, Lepidoptera, Hymenoptera, Diptera, Orthoptera. In future, researchers conduct studies in following aspects, such as constructing public big data sets, minimizing subjective impact photography, delving into interpretable DL, increasing study diverse species. provides new idea development to intervene occurrence pests soon possible. not only reduce chemical pollution but also contribute green earth.

Язык: Английский

---